Fantasy - Deconflicting services at Herne Hill

[squingo44]

The Network Rail 2022 South London Strategy report discusses the same intervention at Herne Hill as I have proposed: see page 70.

It’s in the context of introducing Victoria - Herne Hill - Crystal Palace services to bypass projected capacity constraints at Clapham Junction, and intended to run as parallel movements with Blackfriars - Beckenham Junction slow services. My proposal to introduce a crossover south of Tulse Hill essentially builds on this.

== Doublepost prevention - post automatically merged: 10 Jun 2026 ==

Speculatively doubling transport at vast public expense in the hope the private sector housing market will respond accordingly would be a grossly irresponsible thing to do.

Personally, I feel our planning policy is too risk-averse. We’re limiting our potential exposure to the massive positive payoffs of building new infrastructure, such as the 10x higher usage of the lizzie line than expected. It’s a matter of taste, of course, but there’s a lot of opportunities we’re leaving on the table.

I guess I’m a lot more concerned about achieving economic growth by any means necessary than the rest of the people in this discussion.

 

[35B]

Part of the reason the Victoria and Northern lines are so busy is because at the south end their catchment areas are uncharacteristically large, due to passengers skipping past their local rail service in favour of a high frequency service they don’t have to plan around.

There would be a significant amount of value in creating more rail corridors on which passengers could rely, not least by relieving pressure on the existing tube lines.

Due to their low frequency, most railways in south London do not fulfil this purpose, and they should. There are gaps in the commuter network where there ARE currently trains!

Where you look at crammed commuter services and see a reason not to change what we’ve already got, I see huge amount of unmet demand for transportation.

Thank you. I now see a link between solution and question. I am not convinced by the analysis beyond that, in particular the view of how Tube railheads work or the purpose they serve. And I especially don’t see how the diagnosis of unmet demand relates to the prescribed cure

== Doublepost prevention - post automatically merged: 10 Jun 2026 ==

Personally, I feel our planning policy is too risk-averse. We’re limiting our potential exposure to the massive positive payoffs of building new infrastructure, such as the 10x higher usage of the lizzie line than expected. It’s a matter of taste, of course, but there’s a lot of opportunities we’re leaving on the table.

I guess I’m a lot more concerned about achieving economic growth by any means necessary than the rest of the people in this discussion.

Economic growth needs to serve a purpose, in a society that is inhabited by people rather than units of production. And as for the nature of urban development, I suggest that the factors determining the patterns are much deeper than a "conservative" planning system

 

[Recessio]

In a nutshell, to the rest of London’s public transport network. The current situation is not sufficient because services isn’t frequent enough on the Sutton Loop for passengers to turn and go where they want, and interchange stations don’t exist.

To me, the lack of frequency is the problem statement. Or the fact that users of the south London network have to plan their journeys at all, rather than just hopping on at nearest station at their convenience. When commuters are ignoring their closest station to get the bus to the Northern line, Victoria line, or Windrush line, because the service from their closest station can’t fit more than 2 TPH, that’s a problem.

Part of the reason the Victoria and Northern lines are so busy is because at the south end their catchment areas are uncharacteristically large, due to passengers skipping past their local rail service in favour of a high frequency service they don’t have to plan around.

There would be a significant amount of value in creating more rail corridors on which passengers could rely, not least by relieving pressure on the existing tube lines.

Due to their low frequency, most railways in south London do not fulfil this purpose, and they should. There are gaps in the commuter network where there ARE currently trains!

Where you look at crammed commuter services and see a reason not to change what we’ve already got, I see huge amount of unmet demand for transportation.

I think this situation TfL have identified, where people travel from their nearest railway station to a further tube station, is definitely a problem as it suggests underutilisation of the mainline rail network. As a resident of the Sutton Loop, I see it all the time with commuters travelling to Morden LU instead to get the Northern Line despite the fact many of the most likely final destinationa (e.g. Elephant & Castle, London Bridge, the City, or King's Cross St Pancras) is served by both TL and the Northern Line.

 

[NCT]

I think this situation TfL have identified, where people travel from their nearest railway station to a further tube station, is definitely a problem as it suggests underutilisation of the mainline rail network. As a resident of the Sutton Loop, I see it all the time with commuters travelling to Morden LU instead to get the Northern Line despite the fact many of the most likely final destinationa (e.g. Elephant & Castle, London Bridge, the City, or King's Cross St Pancras) is served by both TL and the Northern Line.

You can reduce 'tube heading' by modest improvements in NR services - operating a base network of 4tph reliably, including say increasing Sutton Loop to 4tph each way. Still some £ and trade-offs will be required, but this is a lot less than flooding the network with double-digit tph and being absolutist about deconflicting routes.

 

[35B]

I think this situation TfL have identified, where people travel from their nearest railway station to a further tube station, is definitely a problem as it suggests underutilisation of the mainline rail network. As a resident of the Sutton Loop, I see it all the time with commuters travelling to Morden LU instead to get the Northern Line despite the fact many of the most likely final destinationa (e.g. Elephant & Castle, London Bridge, the City, or King's Cross St Pancras) is served by both TL and the Northern Line.

Given that duplication, is it not more that getting on the Northern line that early makes the journey to the ultimate destination more comfortable?

 

[Recessio]

Given that duplication, is it not more that getting on the Northern line that early makes the journey to the ultimate destination more comfortable?

Depends if you view riding 20 minutes on a bus in traffic and then crammed onto an un-air-conditioned 95 stock as more comfortable than a direct class 700 from much closer to my house.

I'm lucky at West Sutton that it's effectively 4tph because there's not much difference in journey time for going the "wrong" way around the loop (in fact it's actually quicker to go via Sutton and Mitcham than "direct" via Wimbledon). So for me it makes more sense to take the TL. But at other parts of the loop, the 2tph is not a turn up and go service, discouraging use vs the Northern and Victoria lines.

You can reduce 'tube heading' by modest improvements in NR services - operating a base network of 4tph reliably, including say increasing Sutton Loop to 4tph each way. Still some £ and trade-offs will be required, but this is a lot less than flooding the network with double-digit tph and being absolutist about deconflicting routes.

I've long been in favour of terminating the service at Blackfriars and getting 4tph - changing at Blackfriars or London Bridge is still considerably less faff than taking a bus to Morden - but I understand there are some potential issues with junctions and stock allocations that don't make this quite so simple as it sounds. Still a lot more simpler than the other ideas in this thread, mind you!

 

[35B]

Depends if you view riding 20 minutes on a bus in traffic and then crammed onto an un-air-conditioned 95 stock as more comfortable than a direct class 700 from much closer to my house.

I'm lucky at West Sutton that it's effectively 4tph because there's not much difference in journey time for going the "wrong" way around the loop (in fact it's actually quicker to go via Sutton and Mitcham than "direct" via Wimbledon). So for me it makes more sense to take the TL. But at other parts of the loop, the 2tph is not a turn up and go service, discouraging use vs the Northern and Victoria lines.

I've long been in favour of terminating the service at Blackfriars and getting 4tph - changing at Blackfriars or London Bridge is still considerably less faff than taking a bus to Morden - but I understand there are some potential issues with junctions and stock allocations that don't make this quite so simple as it sounds. Still a lot more simpler than the other ideas in this thread, mind you!

I wouldn't take a bus 20 minutes to get a Northern Line train either but different people, different strokes.

I don't argue with increasing service frequency or have a particular view on whether Sutton trains should or shouldn't go through the Core - again, views of that will vary. But there is a difference between stepping up from 2 tph to 4tph, and the Tube style densities being suggested.

I then also am left with the question of value. Whether or not expressed in money terms, my question is very simple - would delivering these changes deliver a benefit commensurate with the effort required?

 

[cle]

I wouldn't take a bus 20 minutes to get a Northern Line train either but different people, different strokes.

No, but you might jump off at Wimbledon to get the train to Waterloo, which many do.

It’s in the context of introducing Victoria - Herne Hill - Crystal Palace services to bypass projected capacity constraints at Clapham Junction, and intended to run as parallel movements with Blackfriars - Beckenham Junction slow services. My proposal to introduce a crossover south of Tulse Hill essentially builds on this.

I never saw this - but it's interesting.

Victoria-Tulse Hill is unserved, and would be just 3 stops (assuming a Brixton call) - that would be popular, as would the fewer stations for West Norwood and beyond to Victoria. Run it from p7/8 and keep it GTR, it's a decent idea if pathable.

 

[squingo44]

I then also am left with the question of value. Whether or not expressed in money terms, my question is very simple - would delivering these changes deliver a benefit commensurate with the effort required?

I believe so. Admittedly, this is a long term project which should come after many of the other plans in motion or waiting for funding, but at a certain point we're going to need more high-frequncy rail corridors in south London. And this proposal is a cost-effective way of achieving it.

 

[zwk500]

And this proposal is a cost-effective way of achieving it.

You'll need to wait for somebody to calculate an NPV or BCR before making that statement.

 

[squingo44]

Victoria-Tulse Hill is unserved, and would be just 3 stops (assuming a Brixton call) - that would be popular, as would the fewer stations for West Norwood and beyond to Victoria. Run it from p7/8 and keep it GTR, it's a decent idea if pathable.

Another advantage of running Crystal Palace to Victoria as a main passenger rail corridor is that it isolates the interaction of freight rail with the passenger rail network. The parts of the south London network that see rail freight traffic include the South London line, the Catford Loop, the West London Line, the Brighton Main Line relief lines from East Croydon to Clapham Junction, and the lines between Streatham Hill, Tulse Hill, and Herne Hill.

The Dulwich line, Sutton Loop, Herne Hill Viaduct and Chatham Main Line between Herne Hill and Shortlands do not receive any freight traffic, but all run services that do interline with freight traffic in other places. If these sections of track were instead run with services entirely segregated from freight traffic, it would have the potential to reduce operational complexity and improve service frequency.

== Doublepost prevention - post automatically merged: 10 Jun 2026 ==

You'll need to wait for somebody to calculate an NPV or BCR before making that statement.

Since this is all speculative, why can't we eyeball the costs of the various approaches to deconflicting services beyond the metroisation proposal and compare them?

Most other approaches would involve tunnelling of some kind to create a through connection to North London, which is my reason for claiming this approach is comparatively cost-effective. But I'd welcome any other ideas for introducing new passenger rail corridors in south London metro-level frequencies.

 

[Class15]

The Dulwich line, Sutton Loop, Herne Hill Viaduct and Chatham Main Line between Herne Hill and Shortlands do not receive any freight traffic

The Sutton loop has some stone trains towards Tolworth from the Western region.

 

[Park47515]

In a nutshell, to the rest of London’s public transport network. The current situation is not sufficient because services isn’t frequent enough on the Sutton Loop for passengers to turn and go where they want, and interchange stations don’t exist.

Under my proposal, passengers from the loop would have access to a high-frequency service that interchanges with other services to get them pretty much anywhere in London. Change at Streatham Common or Tulse Hill for trains to Victoria. Change at Peckham Rye for the Windrush Line. South Bermondsey for out-of-station interchange with the future Bakerloo Line station at Old Kent Road. And all the interchange opportunities at London Bridge.

Ideally, in the future, such a line would become connected to a commuter line in North London via a through tunnel, but for the moment that doesn’t matter. Joining up the Sutton Loop with the Dulwich line is an effective way of getting >12 TPH service, which is more important than the terminus of the line.

Part of the reason the Victoria and Northern lines are so busy is because at the south end their catchment areas are uncharacteristically large, due to passengers skipping past their local rail service in favour of a high frequency service they don’t have to plan around.

There would be a significant amount of value in creating more rail corridors on which passengers could rely, not least by relieving pressure on the existing tube lines.

Due to their low frequency, most railways in south London do not fulfil this purpose, and they should. There are gaps in the commuter network where there ARE currently trains!

Where you look at crammed commuter services and see a reason not to change what we’ve already got, I see huge amount of unmet demand for transportation.

You'll need to wait for somebody to calculate an NPV or BCR before making that statement.

The BCR will be nearly 0 I would expect

 

[35B]

No, but you might jump off at Wimbledon to get the train to Waterloo, which many do.

Quite likely - but then so might I if it would halve the journey time onwards and/or deliver me to a more convenient part of central London.

 

[NorthKent1989]

View attachment 205681

View attachment 205682

The issue is what currently exists is a low frequency service to many terminals. TfL are quite clear on the benefits a sustained, predictable high-frequency service to one terminal can bring. The case for "One line one terminal" is quite strong. Commuters shouldn't need to check the train times and guess where the most efficient connections can be made just because they live in this area of London, it quite clearly has an inferior public transport system. The issue is any chance for a change and an objectively better service is hushed out because of "historic links" that realise no benefits when it is inconsistent with the rest of the city's public transport system and lead to ridiculous wait times (by a Ldn standard).

Quite frankly nobody dislikes having more connections to multiple terminals, but when it comes at the expense of one sustained, high frequency service then it shouldn't happen.

+ Victoria*

If you need or work in the West End, CX is useful like wise CST is useful for the city, whether you are changing trains at London Bridge for a specific terminal or on a direct service people will always go for the direct service, changing trains more often than not takes time.

 

[squingo44]

If you need or work in the West End, CX is useful like wise CST is useful for the city, whether you are changing trains at London Bridge for a specific terminal or on a direct service people will always go for the direct service, changing trains more often than not takes time.

But 6 TPH to one terminus beats out 2 TPH to two termini

 

[Mattplans]

Agreed.

 

[Magdalia]

I guess I’m a lot more concerned about achieving economic growth by any means necessary than the rest of the people in this discussion.

I studied economics, and worked in economics for 30 years. I am against your proposal because it won't deliver economic growth.

We’re limiting our potential exposure to the massive positive payoffs of building new infrastructure

Not all new infrastructure delivers massive economic payoffs. That only happens if economic growth is constrained by inadequate infrastructure, and the new infrastructure breaks those constraints. When the London economy was growing rapidly, mainly through financial innovation, it needed new infrastructure to grow. The Elizabeth Line is the last in a long line of such projects, delivered many years late.

But it is much harder to make a case that London is constrained by inadequate infrastructure now. Peak time commuting is still about 20% lower than pre-Covid, mainly due to changes in working practices. There will be a further hit to London commuting when artificial intelligence replaces lots of people in "back office functions", what the boss of Standard Chartered recently described as "lower value human capital".

Now the places where inadequate transport infrastructure is constraining economic growth are not in London. The massive payoffs that you are seeking are in the big cities of the north and in the Oxford-Cambridge corridor, not London.

 

[35B]

But 6 TPH to one terminus beats out 2 TPH to two termini

Not necessarily, and it depends on where you measure from. Any analysis that discounts the disruption to a pre-existing journey caused by a newly imposed change is inherently defective. When that change is somewhere like London Bridge, and has the effect of converting a usually seated journey into an almost certainly standing journey, that weighs very heavy indeed.

I am open to the possibility that this might be outweighed by extra gains due to the additional frequency but, as my kids are told by their teachers, the workings must be shown.

 

[squingo44]

converting a usually seated journey into an almost certainly standing journey

I'm unconvinced that this could be the case when the timetable changes will always result in greater capacity, and therefore more seats available.

The only case would be when introducing an interchange to a different service, but since services closer to the centre will certainly have more capacity still, I'm still not convinced that this is an issue. And, even then if the final leg of the journey is spent standing, it's only for a shorter length of time.

With proper segregation of services, I wouldn't expect anyone who starts from an outer commuter service and changes onto another service to complete their commute, to not be able to find a seat on the first service. The only reason that is the case currently is when capacity is very low on a commuter line, usually the 2 TPH services.

 

[Bald Rick]

Capacity IS connectivity.

It isn’t. Improved connectivity is linking places that are not linked now, or drastically improving frequency between places that have poor frequency now.

I’ve never understood why the 5th track is needed and why the trains that use it cant just turn left at the junction instead.

It is so that a train for Elephant & Castle can stand and wait at TVS1079 for a northbound ex Met Jn, and still be clear of following southbound trains towards Met Jn.

 

[squingo44]

drastically improving frequency between places that have poor frequency now.

What else do you think I meant when i said capacity?

 

[NCT]

I've only just got my head around Sutton Loop timings. The current pattern is remarkably neat - the 2tph on each arm make parallel moves at both Blackfriars Junction and Streatham South Junction.

It's tempting to say why not just run Blackfriars Bay to Blackfriars Bay services around the loop 15 minutes from the Thameslink services. Yes, these would also form parallel moves with themselves at Streatham South Junction, but they wouldn't work a Wimbledon.

For 4tph in each direction to be evenly distributed at Wimbledon (theoretically possible I believe), you would indeed need to extend the loop time by about 7 minutes, at the cost of breaking parallel moves at Streatham South Junction.

 

[Recessio]

But 6 TPH to one terminus beats out 2 TPH to two termini

Not necessarily, and it depends on where you measure from. Any analysis that discounts the disruption to a pre-existing journey caused by a newly imposed change is inherently defective. When that change is somewhere like London Bridge, and has the effect of converting a usually seated journey into an almost certainly standing journey, that weighs very heavy indeed.

I am open to the possibility that this might be outweighed by extra gains due to the additional frequency but, as my kids are told by their teachers, the workings must be shown.

6 tph also is a "turn up and go" frequency whereas 2 tph isn't. So there's also quite a strong mental effect for how the service is perceived/how much of an improvement that is, more so than if it was e.g. 6 tph up to 18 tph

 

[Mattplans]

It is so that a train for Elephant & Castle can stand and wait at TVS1079 for a northbound ex Met Jn, and still be clear of following southbound trains towards Met Jn.

The signal in question for anyone wondering:


and why did they not add crossovers so trains from london bridge can reach blackfriars plats 3/4?

 

[Recessio]

I've only just got my head around Sutton Loop timings. The current pattern is remarkably neat - the 2tph on each arm make parallel moves at both Blackfriars Junction and Streatham South Junction.

It's tempting to say why not just run Blackfriars Bay to Blackfriars Bay services around the loop 15 minutes from the Thameslink services. Yes, these would also form parallel moves with themselves at Streatham South Junction, but they wouldn't work a Wimbledon.

For 4tph in each direction to be evenly distributed at Wimbledon (theoretically possible I believe), you would indeed need to extend the loop time by about 7 minutes, at the cost of breaking parallel moves at Streatham South Junction.

Not sure on the timings, but I think you would also need to evict Tramlink from Wimbledon when it's rebuilt for CR2 and restore Platform 10 for use as a second Thameslink platform.

 

[NCT]

Or evict Tramlink from Wimbledon when it's rebuilt for CR2 and restore Platform 10 for use as a second Thameslink platform.

Well, how?

To me the official Crossrail 2's plan of demolishing half of Wimbledon town centre was never going to fly. Crossrail 2 should simply take over the existing slow lines, let existing fast lines be dedicated 'first stop Surbiton' services, and send outer trains (first stop Woking) into a tunnel between Berrylands and Clapham Junction. This way you avoid doing any work at Wimbledon at all.

 

[Bald Rick]

What else do you think I meant when i said capacity?

You said connectivity…

and why did they not add crossovers so trains from london bridge can reach blackfriars plats 3/4?

because they wouldn’t fit along with what was else was more important.

Well, how?

To me the official Crossrail 2's plan of demolishing half of Wimbledon town centre was never going to fly. Crossrail 2 should simply take over the existing slow lines, let existing fast lines be dedicated 'first stop Surbiton' services, and send outer trains (first stop Woking) into a tunnel between Berrylands and Clapham Junction. This way you avoid doing any work at Wimbledon at all.

That was looked at. It still needed wotk at Wimbledon (lots) and access shafts, and buildign a portsl t Clapham was nigh on impossible. Overall it was more expensive.

Besides, Wimbledon were quite keen on losing the shopping centre.

 

[Mattplans]

I am open to the possibility that this might be outweighed by extra gains due to the additional frequency but, as my kids are told by their teachers, the workings must be shown.

Post #148 is pretty convincing.

Any analysis that discounts the disruption to a pre-existing journey caused by a newly imposed change is inherently defective. When that change is somewhere like London Bridge, and has the effect of converting a usually seated journey into an almost certainly standing journey, that weighs very heavy indeed.

We're speaking a lot in hypotheticals, perhaps we could be a little bit more direct about which services specifically you are referring to (my bold). As my firsthand experience on the SE side is both services that run from Victoria and London Bridge towards Lewisham are full and standing. For Victoria, as soon as the Gravesend(via Lewisham) service leaves, the next train leaving 30m later already begins to fill. As for services from London Bridge, unless you join at Charing Cross or Cannon Street there is a very slim chance of getting a seat on peak trains.

I'm not sure that any plan to rationalise services to one terminal could actually result in more spacious carriages and seats, but it would definitely cut out ridiculous extended waits

• 8.1.1 Metroisation offers a substantial improvement in the predictability of rail services in south and south east London. Much like the Tube, customers could expect to turn up any day of the week at their station and get frequent trains to consistent destinations.
• 8.1.2 This predictability would build trust, encouraging more customers to use the services. This confidence in public transport could lead weekday commuters to also consider rail for journeys at the weekend which they would otherwise have driven.

 

[NCT]

That was looked at. It still needed wotk at Wimbledon (lots) and access shafts, and buildign a portsl t Clapham was nigh on impossible. Overall it was more expensive.

Besides, Wimbledon were quite keen on losing the shopping centre.

I believe it was only (more) expensive because even with the fast line tunnels they insisted on keeping 6-tracking Wimbledon - Raynes Park and new terminating platforms at Wimbledon, in order that the SW branches kept their Waterloo trains. My CR2 would have none of that - all suburban services from Epsom, Chessington, Hampton Court and Kingston become CR2, for Waterloo change at Wimbledon (cross-platform) for semi-fast services); and there'd be no Tooting.

As for portals at Clapham Junction - my idea would be for a double set of portals (CR2 dives down into its own tunnels from Earlesfield, and new fast lines emerging from the tunnels to the existing adjacent to Wandsworth cemetery / Wandsworth prison. There's 1.2km to play with between Earlsfield Station and Trinity Way.